Evaluating deep learning architectures for Speech Emotion Recognition

Speech Emotion Recognition (SER) can be regarded as a static or dynamic classification problem, which makes SER an excellent test bed for investigating and comparing various deep learning architectures. We describe a frame-based formulation to SER that relies on minimal speech processing and end-to-...

Full description

Saved in:
Bibliographic Details
Published inNeural networks Vol. 92; pp. 60 - 68
Main Authors Fayek, Haytham M., Lech, Margaret, Cavedon, Lawrence
Format Journal Article
LanguageEnglish
Published United States Elsevier Ltd 01.08.2017
Subjects
Affective computing
Deep learning
Emotion recognition
Emotions
Machine Learning
Neural networks
Neural Networks (Computer)
Speech recognition
Speech Recognition Software
Deep learning
Affective computing
Emotion recognition
Neural networks
Speech recognition
Online AccessGet full text

Cover

More Information
Summary:Speech Emotion Recognition (SER) can be regarded as a static or dynamic classification problem, which makes SER an excellent test bed for investigating and comparing various deep learning architectures. We describe a frame-based formulation to SER that relies on minimal speech processing and end-to-end deep learning to model intra-utterance dynamics. We use the proposed SER system to empirically explore feed-forward and recurrent neural network architectures and their variants. Experiments conducted illuminate the advantages and limitations of these architectures in paralinguistic speech recognition and emotion recognition in particular. As a result of our exploration, we report state-of-the-art results on the IEMOCAP database for speaker-independent SER and present quantitative and qualitative assessments of the models’ performances.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:0893-6080
1879-2782
DOI:10.1016/j.neunet.2017.02.013